4
* Copyright (C) 2009 to 2011 Tim Marston <tim@ed.am>
4
* Copyright (C) 2009 to 2013 Tim Marston <tim@ed.am>
6
6
* This file is part of the Import Contacts program (hereafter referred
7
* to as "this program"). For more information, see
7
* to as "this program"). For more information, see
8
8
* http://ed.am/dev/android/import-contacts
10
10
* This program is free software: you can redistribute it and/or modify
45
45
import java.util.regex.Matcher;
46
46
import java.util.regex.Pattern;
48
import android.annotation.SuppressLint;
49
48
import android.content.SharedPreferences;
49
import android.os.Environment;
51
51
public class VcardImporter extends Importer
71
70
File[] files = null;
73
// check SD card is mounted
74
String state = Environment.getExternalStorageState();
75
if( !Environment.MEDIA_MOUNTED.equals( state ) &&
76
!Environment.MEDIA_MOUNTED_READ_ONLY.equals( state ) )
78
showError( R.string.error_nosdcard );
75
String path = "/sdcard" + prefs.getString( "location", "/" );
76
File file = new File( path );
82
File file = new File( Environment.getExternalStorageDirectory(),
83
prefs.getString( "location", "/" ) );
77
84
if( !file.exists() )
78
85
showError( R.string.error_locationnotfound );
133
140
boolean in_vcard = false;
134
141
while( ( line = reader.readLine() ) != null )
137
145
// look for vcard beginning
138
if( line.matches( "^BEGIN:VCARD" ) ) {
146
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
150
// check for vMsg files
151
else if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VMSG.*" ) ) {
152
showError( getText( R.string.error_vcf_vmsgfile )
143
else if( line.matches( "^END:VCARD" ) )
156
else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
144
157
in_vcard = false;
148
162
catch( FileNotFoundException e ) {
170
184
FileInputStream istream = new FileInputStream( file );
171
185
byte[] content = new byte[ (int)file.length() ];
172
186
istream.read( content );
176
190
importVCardFileContent( content, file.getName() );
192
catch( OutOfMemoryError e ) {
193
showError( R.string.error_outofmemory );
178
195
catch( FileNotFoundException e ) {
179
196
showError( getText( R.string.error_filenotfound ) +
180
197
file.getName() );
193
210
ContentLineIterator cli = new ContentLineIterator( content );
194
211
while( cli.hasNext() )
196
ByteBuffer buffer = cli.next();
213
ContentLine content_line = cli.next();
198
// get a US-ASCII version of the line for processing
201
line = new String( buffer.array(), buffer.position(),
202
buffer.limit() - buffer.position(), "US-ASCII" );
204
catch( UnsupportedEncodingException e ) {
205
// we know US-ASCII is supported, so appease the compiler...
215
// get a US-ASCII version of the string, for processing
216
String line = content_line.getUsAsciiLine();
209
218
if( vcard == null ) {
210
219
// look for vcard beginning
211
if( line.matches( "^BEGIN:VCARD" ) ) {
220
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
212
221
setProgress( _progress++ );
213
222
vcard = new Vcard();
214
223
vcard_start_line = cli.getLineNumber();
260
269
// try giving the line to the vcard
262
vcard.parseLine( buffer, line,
263
cli.doesNextLineLookFolded() );
271
vcard.parseLine( content_line );
265
273
catch( Vcard.ParseException e ) {
273
281
finish( ACTION_ABORT );
276
// although we're continuing, we still need to abort
277
// this vCard. Further lines will be ignored until we
284
// Although we're continuing, we still need to abort
285
// this vCard. Further lines will be ignored until we
278
286
// get to another BEGIN:VCARD line.
281
289
catch( Vcard.SkipImportException e ) {
283
// abort this vCard. Further lines will be ignored until
291
// Abort this vCard. Further lines will be ignored until
284
292
// we get to another BEGIN:VCARD line.
292
class ContentLineIterator implements Iterator< ByteBuffer >
302
private ByteBuffer _buffer;
303
private boolean _folded_next;
304
private String _line;
306
public ContentLine( ByteBuffer buffer, boolean folded_next )
309
_folded_next = folded_next;
313
public ByteBuffer getBuffer()
318
public boolean doesNextLineLookFolded()
323
public String getUsAsciiLine()
325
// generated line and cache it
326
if( _line == null ) {
328
_line = new String( _buffer.array(), _buffer.position(),
329
_buffer.limit() - _buffer.position(), "US-ASCII" );
331
catch( UnsupportedEncodingException e ) {
332
// we know US-ASCII *is* supported, so appease the
337
// return cached line
342
class ContentLineIterator implements Iterator< ContentLine >
294
344
protected byte[] _content = null;
295
345
protected int _pos = 0;
320
370
_pos > initial_pos )? _pos - 1 : _pos;
323
return ByteBuffer.wrap( _content, initial_pos,
373
return new ContentLine(
374
ByteBuffer.wrap( _content, initial_pos,
376
doesNextLineLookFolded() );
327
379
// we didn't find one, but were there bytes left?
332
return ByteBuffer.wrap( _content, initial_pos,
384
return new ContentLine(
385
ByteBuffer.wrap( _content, initial_pos,
387
doesNextLineLookFolded() );
348
402
* onto the end of this one?
351
public boolean doesNextLineLookFolded()
405
private boolean doesNextLineLookFolded()
353
407
return _pos > 0 && _pos < _content.length &&
354
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
408
_content[ _pos - 1 ] == '\n' &&
409
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
357
412
public int getLineNumber()
369
424
private final static int MULTILINE_NONE = 0;
370
425
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
371
426
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
372
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
427
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
374
429
private String _version = null;
375
private Vector< ByteBuffer > _buffers = null;
430
private Vector< ContentLine > _content_lines = null;
376
431
private int _name_level = NAMELEVEL_NONE;
377
432
private int _parser_multiline_state = MULTILINE_NONE;
378
433
private String _parser_current_name_and_params = null;
421
476
@SuppressWarnings("serial")
422
477
protected class SkipImportException extends Exception { }
424
private String extractCollonPartFromLine( ByteBuffer buffer,
425
String line, boolean former )
479
private String extractCollonPartFromLine( ContentLine content_line,
429
// get a US-ASCII version of the line for processing, unless we were
433
line = new String( buffer.array(), buffer.position(),
434
buffer.limit() - buffer.position(), "US-ASCII" );
436
catch( UnsupportedEncodingException e ) {
437
// we know US-ASCII is supported, so appease the compiler...
442
482
// split line into name and value parts and check to make sure we
443
483
// only got 2 parts and that the first part is not zero in length
444
String[] parts = line.split( ":", 2 );
484
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
445
485
if( parts.length == 2 && parts[ 0 ].length() > 0 )
446
ret = parts[ former? 0 : 1 ];
451
private String extractNameAndParamsFromLine( ByteBuffer buffer,
454
return extractCollonPartFromLine( buffer, line, true );
457
private String extractValueFromLine( ByteBuffer buffer, String line )
459
return extractCollonPartFromLine( buffer, line, false );
462
public void parseLine( ByteBuffer buffer, String line,
463
boolean next_line_looks_folded )
486
return parts[ former? 0 : 1 ].trim();
491
private String extractNameAndParamsFromLine( ContentLine content_line )
493
return extractCollonPartFromLine( content_line, true );
496
private String extractValueFromLine( ContentLine content_line )
498
return extractCollonPartFromLine( content_line, false );
501
public void parseLine( ContentLine content_line )
464
502
throws ParseException, SkipImportException,
465
503
AbortImportException
470
508
// tentatively get name and params from line
471
509
String name_and_params =
472
extractNameAndParamsFromLine( buffer, line );
510
extractNameAndParamsFromLine( content_line );
474
512
// is it a version line?
475
513
if( name_and_params != null &&
476
name_and_params.equals( "VERSION" ) )
514
name_and_params.equalsIgnoreCase( "VERSION" ) )
479
String value = extractValueFromLine( buffer, line );
480
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
517
String value = extractValueFromLine( content_line );
518
if( value == null || (
519
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
481
521
throw new ParseException( R.string.error_vcf_version );
482
523
_version = value;
484
525
// parse any buffers we've been accumulating while we waited
486
if( _buffers != null )
487
for( int i = 0; i < _buffers.size(); i++ )
488
parseLine( _buffers.get( i ), null,
489
i + 1 < _buffers.size() &&
490
_buffers.get( i + 1 ).hasRemaining() &&
491
_buffers.get( i + 1 ).get(
492
_buffers.get( i + 1 ).position() ) == ' ' );
527
if( _content_lines != null )
528
for( int i = 0; i < _content_lines.size(); i++ )
529
parseLine( _content_lines.get( i ) );
530
_content_lines = null;
497
534
// no, so stash this line till we get a version
498
if( _buffers == null )
499
_buffers = new Vector< ByteBuffer >();
500
_buffers.add( buffer );
535
if( _content_lines == null )
536
_content_lines = new Vector< ContentLine >();
537
_content_lines.add( content_line );
505
542
// name and params and the position in the buffer where the
506
// "value" part of the line start
543
// "value" part of the line starts
507
544
String name_and_params;
516
553
// skip some initial line characters, depending on the type
517
554
// of multi-line we're handling
518
pos = buffer.position();
555
pos = content_line.getBuffer().position();
519
556
switch( _parser_multiline_state )
521
558
case MULTILINE_FOLDED:
524
561
case MULTILINE_ENCODED:
525
while( pos < buffer.limit() && (
526
buffer.get( pos ) == ' ' ||
527
buffer.get( pos ) == '\t' ) )
562
while( pos < content_line.getBuffer().limit() && (
563
content_line.getBuffer().get( pos ) == ' ' ||
564
content_line.getBuffer().get( pos ) == '\t' ) )
580
if( content_line.getUsAsciiLine().trim().length() == 0 )
542
583
// get name and params from line, and since we're not
543
584
// parsing a subsequent line in a multi-line, this should
544
585
// not fail, or it's an error
545
586
name_and_params =
546
extractNameAndParamsFromLine( buffer, line );
587
extractNameAndParamsFromLine( content_line );
547
588
if( name_and_params == null )
548
589
throw new ParseException(
549
590
R.string.error_vcf_malformed );
551
592
// calculate how many chars to skip from beginning of line
552
593
// so we skip the property "name:" part
553
pos = buffer.position() + name_and_params.length() + 1;
594
pos = content_line.getBuffer().position() +
595
name_and_params.length() + 1;
555
597
// reset the saved multi-line state
556
598
_parser_current_name_and_params = name_and_params;
560
602
// get value from buffer, as raw bytes
561
603
ByteBuffer value;
562
value = ByteBuffer.wrap( buffer.array(), pos,
563
buffer.limit() - pos );
604
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
605
content_line.getBuffer().limit() - pos );
565
607
// get parameter parts
566
608
String[] name_param_parts = name_and_params.split( ";", -1 );
573
615
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
575
617
boolean is_interesting_field =
576
interesting_fields.contains( name_param_parts[ 0 ] );
618
interesting_fields.contains(
619
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
578
621
// parse encoding parameter
579
622
String encoding = checkParam( name_param_parts, "ENCODING" );
580
623
if( encoding != null )
581
624
encoding = encoding.toUpperCase( Locale.US );
582
625
if( is_interesting_field && encoding != null &&
583
!encoding.equals( "8BIT" ) &&
584
!encoding.equals( "QUOTED-PRINTABLE" ) )
585
//&& !encoding.equals( "BASE64" ) )
626
!encoding.equalsIgnoreCase( "8BIT" ) &&
627
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
628
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
587
630
throw new ParseException( R.string.error_vcf_encoding );
592
635
if( charset != null )
593
636
charset = charset.toUpperCase( Locale.US );
594
637
if( charset != null &&
595
!charset.equals( "US-ASCII" ) &&
596
!charset.equals( "ASCII" ) &&
597
!charset.equals( "UTF-8" ) )
638
!charset.equalsIgnoreCase( "US-ASCII" ) &&
639
!charset.equalsIgnoreCase( "ASCII" ) &&
640
!charset.equalsIgnoreCase( "UTF-8" ) )
599
642
throw new ParseException( R.string.error_vcf_charset );
602
645
// do unencoding (or default to a fake unencoding result with
603
646
// the raw string)
604
647
UnencodeResult unencoding_result = null;
605
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
648
if( encoding != null &&
649
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
606
651
unencoding_result = unencodeQuotedPrintable( value );
607
// else if( encoding != null && encoding.equals( "BASE64" ) )
653
// else if( encoding != null &&
654
// encoding.equalsIgnoreCase( "BASE64" ) )
608
656
// unencoding_result = unencodeBase64( props[ 1 ], charset );
609
658
if( unencoding_result != null ) {
610
659
value = unencoding_result.getBuffer();
611
660
if( unencoding_result.isAnotherLineRequired() )
616
665
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
617
666
if( ( charset == null && _version.equals( "2.1" ) ) ||
618
667
( charset != null && (
619
charset.equals( "ASCII" ) ||
620
charset.equals( "US-ASCII" ) ) ) )
668
charset.equalsIgnoreCase( "ASCII" ) ||
669
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
622
671
value = transcodeAsciiToUtf8( value );
634
683
// for some entries that have semicolon-separated value parts,
635
684
// check to see if the value ends in an escape character, which
636
685
// indicates that we have a multi-line value
637
if( ( name_param_parts[ 0 ].equals( "N" ) ||
638
name_param_parts[ 0 ].equals( "ORG" ) ||
639
name_param_parts[ 0 ].equals( "ADR" ) ) &&
686
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
687
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
688
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
640
689
doesStringEndInAnEscapeChar( string_value ) )
642
691
_parser_multiline_state = MULTILINE_ESCAPED;
644
693
string_value.length() - 1 );
647
// now we know whether we're in an encoding multi-line,
648
// determine if we're in a v3 folded multi-line or not
696
// if we know we're not in an encoding-based multi-line, check
697
// to see if we're in a folded multi-line
649
698
if( _parser_multiline_state == MULTILINE_NONE &&
650
_version.equals( "3.0" ) && next_line_looks_folded )
699
content_line.doesNextLineLookFolded() )
652
701
_parser_multiline_state = MULTILINE_FOLDED;
665
714
if( complete_value.length() < 1 ) return;
667
716
// parse some properties
668
if( name_param_parts[ 0 ].equals( "N" ) )
717
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
669
718
parseN( name_param_parts, complete_value );
670
else if( name_param_parts[ 0 ].equals( "FN" ) )
719
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
671
720
parseFN( name_param_parts, complete_value );
672
else if( name_param_parts[ 0 ].equals( "ORG" ) )
721
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
673
722
parseORG( name_param_parts, complete_value );
674
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
723
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
675
724
parseTITLE( name_param_parts, complete_value );
676
else if( name_param_parts[ 0 ].equals( "TEL" ) )
725
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
677
726
parseTEL( name_param_parts, complete_value );
678
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
727
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
679
728
parseEMAIL( name_param_parts, complete_value );
680
else if( name_param_parts[ 0 ].equals( "ADR" ) )
729
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
681
730
parseADR( name_param_parts, complete_value );
682
else if( name_param_parts[ 0 ].equals( "LABEL" ) )
731
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
683
732
parseLABEL( name_param_parts, complete_value );
684
else if( name_param_parts[ 0 ].equals( "NOTE" ) )
733
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
685
734
parseNOTE( name_param_parts, complete_value );
735
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
736
parseBDAY( name_param_parts, complete_value );
713
764
String str = parts.get( a );
715
// look for parts that end in an escape character, but ignore
716
// the final part. We've already detected escape chars at the
766
// Look for parts that end in an escape character, but ignore
767
// the final part. We've already detected escape chars at the
717
768
// end of the final part in parseLine() and handled multi-lines
719
770
if( a < parts.size() - 1 &&
809
860
for( int b = 0; b < name_part_parts.length; b++ )
810
861
if( name_part_parts[ b ].length() > 0 )
812
if( value.length() == 0 ) value += " ";
863
if( value.length() > 0 ) value += " ";
813
864
value += name_part_parts[ b ];
934
985
for( int a = 0; a < adr_parts.length; a++ )
935
986
if( adr_parts[ a ].length() > 0 )
937
// split this part in to it's comma-separated bits
938
String[] adr_part_parts =
939
splitValueByCharacter( adr_parts[ a ], ',' );
940
for( int b = 0; b < adr_part_parts.length; b++ )
941
if( adr_part_parts[ b ].length() > 0 )
943
if( value.length() > 0 ) value += "\n";
944
value += adr_part_parts[ b ];
988
// version 3.0 vCards allow further splitting by comma
989
if( _version.equals( "3.0" ) )
991
// split this part in to it's comma-separated bits and
992
// add them on individual lines
993
String[] adr_part_parts =
994
splitValueByCharacter( adr_parts[ a ], ',' );
995
for( int b = 0; b < adr_part_parts.length; b++ )
996
if( adr_part_parts[ b ].length() > 0 )
998
if( value.length() > 0 ) value += "\n";
999
value += adr_part_parts[ b ];
1004
// add this part on an individual line
1005
if( value.length() > 0 ) value += "\n";
1006
value += adr_parts[ a ];
948
1010
Set< String > types = extractTypes( params, Arrays.asList(
978
1040
addNote( unescapeValue( value ) );
1043
private void parseBDAY( String[] params, String value )
1045
setBirthday( value );
981
1048
public void finaliseVcard()
982
1049
throws ParseException, ContactNotIdentifiableException
984
1051
// missing version (and data is present)
985
if( _version == null && _buffers != null )
1052
if( _version == null && _content_lines != null )
986
1053
throw new ParseException( R.string.error_vcf_malformed );
988
1055
// finalise the parent class
1013
1082
HashSet< String > ret = new HashSet< String >();
1015
1084
Pattern p = Pattern.compile(
1016
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1085
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1086
Pattern.CASE_INSENSITIVE );
1017
1087
for( int i = 0; i < params.length; i++ ) {
1018
1088
Matcher m = p.matcher( params[ i ] );
1019
1089
if( m.matches() )
1027
* Amongst the params, return any type values present. For v2.1 vCards,
1028
* those types are just parameters. For v3.0, they are prefixed with
1029
* "TYPE=". There may also be multiple type parameters.
1031
* @param a list of type values to look for
1097
* Amongst the params, return any type values present. For v2.1 vCards,
1098
* those types are just parameters. For v3.0, they are prefixed with
1099
* "TYPE=". There may also be multiple type parameters.
1101
* @param params an array of params to look for types in
1102
* @param valid_types an list of upper-case type values to look for
1032
1103
* @return a set of present type values
1034
1105
private Set< String > extractTypes( String[] params,
1040
1111
String type_params[] = checkParams( params, "TYPE" );
1041
1112
for( int a = 0; a < type_params.length; a++ )
1043
// check for a comma-separated list of types (why? this isn't in
1114
// check for a comma-separated list of types (why? I don't think
1115
// this is in the specs!)
1045
1116
String[] parts = type_params[ a ].split( "," );
1046
for( int i = 0; i < parts.length; i++ )
1047
if( valid_types.contains( parts[ i ] ) )
1048
types.add( parts[ i ] );
1117
for( int i = 0; i < parts.length; i++ ) {
1118
String ucpart = parts[ i ].toUpperCase( Locale.US );
1119
if( valid_types.contains( ucpart ) )
1120
types.add( ucpart );
1051
1124
// get 2.1-style type param
1052
1125
if( _version.equals( "2.1" ) ) {
1053
for( int i = 1; i < params.length; i++ )
1054
if( valid_types.contains( params[ i ] ) )
1055
types.add( params[ i ] );
1126
for( int i = 1; i < params.length; i++ ) {
1127
String ucparam = params[ i ].toUpperCase( Locale.US );
1128
if( valid_types.contains( ucparam ) )
1129
types.add( ucparam );
1080
1155
else if( ch == '=' && i == in.limit() - 1 )
1082
1157
// we found a '=' at the end of a line signifying a multi-
1083
// line string, so we don't add it.
1158
// line string, so we don't add it
1084
1159
another = true;